Our previous studies have failed to demonstrate a relation between numbers of pulmonary asbestos bodies and the presence of carcinoma of the lung or gastrointestinal tract in workers with very low asbestos exposure. Asbestos bodies, however, represent only a part, and perhaps only a small part, of the total lung asbestos burden, which is composed of coated bodies as well as uncoated fibers. It remains possible that a relation exists between numbers of uncoated asbestos fibers and the presence of bronchogenic carcinoma. This study will attempt to quantify and type the uncoated asbestos fibers as well as asbestos bodies at various sites in the lung. A new preparative method has already been developed and used for examining and identifying the nature of the cores of asbestos bodies by electron diffraction. The same method will be used to identify further these same bodies by electron microprobe analysis. An additional preparative method has been devised for recovering uncoated asbestos fibers. Relations between numbers of coated and uncoated asbestos fibers, as well as their relative locations within the lung, will be tabulated, and an attempt will be made to correlate numbers of types of fibers with the presence or absence of carcinomas. We have previously shown that numbers of asbestos bodies are related to occupation; an attempt will be made to establish whether a similar situation exists for uncoated fibers or whether they are derived from the environment.